Recombinant anti-erbB2 immunotoxins containing Pseudomonas exotoxin.

Immunotoxins were made using five different murine monoclonal antibodies to the human erbB2 gene product and LysPE40, a 40-kDa recombinant form of Pseudomonas exotoxin (PE) lacking its cell-binding domain. All five conjugates were specifically cytotoxic to cancer cell lines overexpressing erbB2 protein. The most active conjugate was e23-LysPE40, generated by chemical crosslinking of anti-erbB2 monoclonal antibody e23 to LysPE40. In addition, a recombinant immunotoxin, e23(Fv)PE40, was constructed that consists of the light-chain variable domain of e23 connected through a peptide linker to its heavy-chain variable domain, which in turn is fused to PE40. The recombinant protein was made in Escherichia coli, purified to near homogeneity, and shown to selectively kill cells expressing the erbB2 protooncogene. To improve the cytotoxic activity of e23(Fv)PE40, PE40 was replaced with a variant, PE38KDEL, in which the carboxyl end of PE is changed from Arg-Glu-Asp-Leu-Lys to Lys-Asp-Glu-Leu and amino acids 365-380 of PE are deleted. The e23(Fv)PE38KDEL protein inhibits the growth of tumors formed by the human gastric cancer cell line N87 in immunodeficient mice.     

Synergy of ricin A chain-containing immunotoxins and ricin B chain-containing immunotoxins in in vitro killing of neoplastic human B cells.

A strategy is described for improving the efficacy of ricin A chain-containing immunotoxins. Highly purified preparations of ricin A chain or ricin B chain were separately coupled to anti-human immunoglobulin antibodies and the conjugates (immunotoxins) were affinity purified to eliminate free chains. Mixtures of anti-Ig-A and anti-Ig-B immunotoxins markedly synergized in vitro in their ability to kill the Ig-bearing human lymphoma cell line Daudi. In contrast, A-chain- or B-chain-containing immunotoxins of irrelevant specificity did not synergize with anti-Ig-A or anti-Ig-B immunotoxins. This finding indicates that free A or B chains do not play a major role in the synergy and that the synergy is specific. Thus, synergy depends on the specificity of the two antibodies; the lectin binding ability of a B-chain-containing immunotoxin of irrelevant specificity does not suffice. This approach of delivering ricin A chain and B chain separately to a target cell may have significant advantages in killing cells that are not effectively killed by A-chain-containing immunotoxins alone.     

Effect of immunotoxins against Trypanosoma cruzi

Immunotoxins were constructed with IgG antibodies against Trypanosoma cruzi surface antigens by hybridization with abrin (ITA) and ricin (ITR) A chains. The biological activity of the hybrid macromolecules was tested on the parasite forms. Motility of parasite forms was lost in vitro after incubation with ITR. In general, killing of the parasite with ITR was more efficient than with ITA. Inhibition of protein synthesis after incubation with either ITR or ITA, measured by 3H-leucine incorporation, confirmed the parasite immobilization experiments. The lethal effect was potentiated when the immunotoxins were used in the presence of 2.5 mM ammonium chloride. T. cruzi antibodies specific to cell surface antigens are excellent drug carriers that can be delivered to the target cell. However, ITR and ITA did not reduce parasitemia or increase survival of mice infected with T. cruzi.     

Humanization of murine monoclonal antibodies through variable domain resurfacing.

Two murine monoclonal antibodies, N901 (anti-CD56) and anti-B4 (anti-CD19), were humanized by a process we call "resurfacing." A systematic analysis of known antibody structures has been used to determine the relative solvent accessibility distributions of amino acid residues in murine and human antibody variable (Fv) regions and has shown that the sequence alignment positions of surface amino acids for human and murine variable region heavy (VH) and light (VL) chains are conserved with 98% fidelity across species. While the amino acid usage at these surface positions creates surface residue patterns that are conserved within species, there are no identical patterns across species. However, surprisingly few amino acid changes need to be made to convert a murine Fv surface pattern to that characteristic of a human surface. Resurfacing was used to change the patterns of surface accessible residues in the Fv regions of the N901 and anti-B4 antibodies to resemble those found on the Fv regions of human antibody sequences. Two different procedures for selecting a human sequence were compared. For anti-B4, a data base of clonally derived human VL-VH sequence pairs was used, while for N901, sequences for VL and VH were independently selected from the Kabat et al. data base [Kabat, E. A., Wu, T. T., Reid-Miller, M., Perry, H. M. & Gottesman, K. S. (1991) Sequences of Proteins of Immunological Interest (DHHS, Washington, DC), 5th Ed.]. Resurfaced N901 and anti-B4 antibodies had apparent affinities for their cell surface ligands that were identical to those of their respective parent murine antibodies. These data provide evidence that, despite the differences in the surfaces of mouse and human Fv regions, it is possible to substitute one for the other while retaining full antigen binding affinity.     

Independent domain folding of Pseudomonas exotoxin and single-chain immunotoxins: influence of interdomain connections.

We have studied the refolding of completely unfolded and reduced Pseudomonas exotoxin (PE) and of recombinant single-chain immunotoxins made with monoclonal antibody B3 that are composed of a heavy-chain variable region connected by a flexible linker to the corresponding light-chain variable region (Fv), which is in turn fused to a truncated form of PE. We have found by direct activity assays that different functional domains of these multifunctional proteins fold independently with different kinetics. The ADP-ribosylation domain of PE and of the recombinant immunotoxin fold rapidly, whereas the assembly of the binding and/or translocation domains is regained more slowly. The complete refolding of native PE occurs more rapidly than the refolding of the recombinant immunotoxins. To determine the influence of the connector region between the B3(Fv) moiety and the toxin on the folding process of the recombinant immunotoxin B3(Fv)-PE38KDEL, we have made two different mutations in the peptide that connects the single-chain Fv domain to domain II of PE. These molecules show different folding kinetics, differences in their propensity to aggregate, and different yields of correctly folded molecules. A mutation that decreases aggregation increases the rate of formation and the yield of active immunotoxin molecules.     

Selective killing of HIV-infected cells by anti-gp120 immunotoxins

Either ricin A chain (RAC) or Pseudomonas exotoxin (PE) was conjugated with a murine monoclonal antibody (0.5 beta) directed against an external envelope glycoprotein (gp120) of human immunodeficiency virus (HIV). Effects of the immunotoxins produced against infected cells were evaluated. Selective inhibition of the proliferation and killing of chronically HIV infected cells were observed in the presence of the immunotoxins. To determine the feasibility of the immunotoxins against the infected cells in seropositive subjects, we attempted to detect gp120-bearing cells in peripheral blood mononuclear cells (PBM) by cytofluorography. Cells in the monocyte/macrophage region of 2 of 10 PBM samples from HIV-infected individuals were found to react with 0.5 beta (18.1% and 12.8%). Furthermore, the cell population which was reactive with 0.5 beta was also susceptible to RAC conjugated with 0.5 beta. These results suggest that the strategy of using anti-gp120 immunotoxin to eliminate HIV-infected cells may be feasible in infected individuals.     

Humanization of an anti-p185HER2 antibody for human cancer therapy.

The murine monoclonal antibody mumAb4D5, directed against human epidermal growth factor receptor 2 (p185HER2), specifically inhibits proliferation of human tumor cells overexpressing p185HER2. However, the efficacy of mumAb4D5 in human cancer therapy is likely to be limited by a human anti-mouse antibody response and lack of effector functions. A "humanized" antibody, humAb4D5-1, containing only the antigen binding loops from mumAb4D5 and human variable region framework residues plus IgG1 constant domains was constructed. Light- and heavy-chain variable regions were simultaneously humanized in one step by "gene conversion mutagenesis" using 311-mer and 361-mer preassembled oligonucleotides, respectively. The humAb4D5-1 variant does not block the proliferation of human breast carcinoma SK-BR-3 cells, which overexpress p185HER2, despite tight antigen binding (Kd = 25 nM). One of seven additional humanized variants designed by molecular modeling (humAb4D5-8) binds the p185HER2 antigen 250-fold and 3-fold more tightly than humAb4D5-1 and mumAb4D5, respectively. In addition, humAb4D5-8 has potency comparable to the murine antibody in blocking SK-BR-3 cell proliferation. Furthermore, humAb4D5-8 is much more efficient in supporting antibody-dependent cellular cytotoxicity against SK-BR-3 cells than mumAb4D5, but it does not efficiently kill WI-38 cells, which express p185HER2 at lower levels.     

Interleukin 2 enhances the efficiency of immunotoxins in the treatment of mice with ascitic tumors.

Two multivalent immunotoxins (ITs) with cytotoxic potential against Thy 1.2-expressing tumor cells were used in association with mouse interleukin 2 (IL2) for treatment of mice bearing ascitic EL4 lymphomas. The combined treatment, ITs + IL2, induced an enhanced antitumor effect revealed by a significant prolongation of the survival time of mice as compared to the simple treatment with ITs or IL2 alone. According to the survival of mice treated by combined therapy, the proportion of killed tumor cells rose up to 94% as resulted from the dose-dependent curve of the survival of nontreated mice versus the number of tumor cells inoculated.     

Relationship between internalization and cytotoxicity of ricin A-chain immunotoxins

Immunotoxins (ITs) appear to vary considerably in their killing efficiency towards antigen positive cells. In order to unravel the mechanisms underlying these differences, the parameters responsible for these variations were studied. The efficacy of the monoclonal antibodies (MoAbs) WT32 (CD3), OKT4 (CD4), T101 (CD5), WT1 (CD7) and WT82 (CD8) conjugated to ricin A-chain was expressed by the extent of protein synthesis inhibition of four leukaemic T cell lines (CEM, GH1, Jurkat and HPB-ALL). Large differences in cytotoxicity were observed. Efficient inhibition of protein synthesis was seen with anti-CD3 IT, anti-CD5 IT and anti-CD7 IT. In these cases the cytotoxicity was related to the antigen density on the target cell membrane. Anti-CD4 IT inhibited poorly and anti-CD8 IT was ineffective, even on cell lines with a high expression of the corresponding antigen. When antigen density and cytotoxicity were plotted for all CD antigens, no correlation could be found. Subsequently, internalization was studied with 125Iodine-labelled antibodies. Anti-CD7 showed the fastest internalization rate, followed by anti-CD3 and anti-CD5. Anti-CD4 and anti-CD8 antibodies were respectively moderately and hardly internalized. When the absolute amount of internalized MoAb was calculated, a highly significant correlation with cytotoxicity was found. We conclude that the degree of antigen expression is not so important as the absolute amount of antibody internalized in predicting the efficacy of ITs.     

A rapid method of cloning functional variable-region antibody genes in Escherichia coli as single-chain immunotoxins.

We have devised a strategy based on polymerase chain reaction (PCR) for the rapid cloning of functional antibody genes as single-chain immunotoxins. RNA from a hybridoma producing an antibody (OVB3) that reacts with ovarian cancer cells was used as a template to make the first strand of a cDNA. Then a second strand was synthesized and amplified by using two sets of DNA primers that (i) hybridized to the ends of the light- and heavy-chain variable regions, (ii) encoded a linker peptide, and (iii) contained appropriate restriction enzyme sites for cloning. After 30 cycles of PCR, the DNA fragments containing sequences encoding the light- and heavy-chain variable regions were cloned into an Escherichia coli expression vector containing a portion of the Pseudomonas exotoxin gene. Clones encoding recombinant single-chain immunotoxins were expressed in E. coli and the protein product was assessed for its ability to bind to or kill cells bearing the OVB3 antigen. By using this approach it should be possible to rapidly clone the functional variable region sequences of many different antibodies from hybridoma RNA.     

Anti-tumor activities of immunotoxins made of monoclonal antibody B3 and various forms of Pseudomonas exotoxin.

B3 is a monoclonal antibody that reacts with a carbohydrate epitope present on a variety of proteins located on the surface of many cancer cells and a limited number of normal tissues. We evaluated the cytotoxic activity of immunotoxins composed of monoclonal antibody B3 coupled to native Pseudomonas exotoxin (PE) or two recombinant forms of Pseudomonas exotoxin, PEArg57 or LysPE40, a form of PE with a deletion of the cell binding domain. All three conjugates were cytotoxic to human cell lines expressing the B3 antigen on their surface. The survival of each of the three immunotoxins in the circulation of mice was determined after administering the immunotoxin i.v. The half-life in blood of B3-PE and B3-PEArg57 was 20 hr, whereas the half-life of B3-LysPE40 was 4 hr. The short half-life of B3-LysPE40 may be due to the absence of domain I of PE. To determine the therapeutic effects of the three immunotoxins, they were given intraperitoneally to nude mice bearing subcutaneous A431 tumors. All three immunotoxins caused complete regression of 50-mm3 tumors with no toxic effects to the animals at therapeutic doses. Furthermore, substantial regression was also noted with much larger tumors. Our data indicate that the monoclonal antibody B3, when coupled to PE or recombinant forms of PE, may be useful for the treatment of tumors expressing B3 antigen. The therapeutic window was largest with B3-LysPE40, which can be administered in higher doses because it lacks sequences in domain I of PE that enable PE to bind to nontarget cells.     

Complete suppression of in vivo growth of human leukemia cells by specific immunotoxins: nude mouse models.

In this study, immunotoxins containing monoclonal anti-human T-cell leukemia antibodies are shown to be capable of completely suppressing the tumor growth of human T-cell leukemia cells in vivo without any overt undesirable toxicity. These immunotoxins were prepared by conjugating ricin A chain (RA) with our monoclonal antibodies, SN1 and SN2, directed specifically to the human T-cell leukemia cell surface antigens TALLA and GP37, respectively. We have shown that these monoclonal antibodies are highly specific for human T-cell leukemia cells and do not react with various normal cells including normal T and B cells, thymocytes, and bone marrow cells. Ascitic and solid human T-cell leukemia cell tumors were generated in nude mice. The ascitic tumor was generated by transplanting Ichikawa cells (a human T-cell leukemia cell line) i.p. into nude mice, whereas the solid tumor was generated by transplanting s.c. MOLT-4 cells (a human T-cell leukemia cell line) and x-irradiated human fibrosarcoma cells into x-irradiated nude mice. To investigate the efficacy of specific immunotoxins in suppressing the in vivo growth of the ascitic tumor, we divided 40 nude mice that were injected with Ichikawa cells into four groups. Each group of 10 mice was injected with one of the following mixtures: 40 micrograms of purified control mouse IgG [IgG1(kappa)] (group 1), 40 micrograms of control RA conjugate (group 2), 20 micrograms of purified SN1 antibody [IgG1(kappa)] and 20 micrograms of purified SN2 antibody [IgG1(kappa)] (group 3), or 20 micrograms of SN1-RA and 20 micrograms of SN2-RA (group 4). Mice in groups 1 and 2 formed large ascitic tumors, and died 5.8-7.0 weeks after the transplantation. Group 3 mice also formed large ascitic tumors and died 6.4-7.8 weeks after the transplantation. However, none of the mice in group 4 that were treated with SN1-RA and SN2-RA showed any signs of a tumor or undesirable toxic effects for the 20 weeks that they were followed after the transplantation; these mice were indistinguishable from healthy control nude mice that were not injected with Ichikawa cells. Treatment with SN1-RA plus SN2-RA completely suppressed solid tumor growth in 4 of 10 nude mice carrying solid tumors and partially suppressed the tumor growth in the remaining 6 nude mice. These results strongly suggest that SN1-RA and SN2-RA may be useful for clinical treatment.     

Streptavidin-biotin immunotoxins: a new approach to purging bone marrow

Immunotoxins have been used both experimentally and clinically to purge bone marrow of tumor cells or T cells before transplantation. We describe the synthesis of a streptavidin-biotin-toxin conjugate using whole ricin. Streptavidin-biotin-ricin (SA-BR) conjugates were synthesized by biotinylation of whole ricin, which was then complexed with streptavidin. Hybrid molecules consisting of a single biotinylated ricin moiety linked to a streptavidin molecule were separated by gel filtration. This SA-BR conjugate was used in an indirect cytotoxicity assay. The assay involved sensitizing of target cells with biotinylated monoclonal antibody (B-MCAB) followed by treatment with dilutions of SA-BR conjugate. The assay demonstrated a specific antibody-directed cytotoxicity. The strength of this SA-BR system is that a single conjugate was able to be used in conjunction with a library of B-MCABs to selectively target phenotypically different cell types. The application of the SA-BR conjugate is thus only restricted by the availability of B-MCABs specific for the desired target cells. The high affinity of avidin for biotin (Kd approximately 10(-15)) and the ability of a single conjugate to target phenotypically different cells through utilization of a library of B-MCABs gives SA-BR conjugates great potential in the selective targeting of individual cell types.     

Soluble CD4 enhances the efficacy of immunotoxins directed against gp41 of the human immunodeficiency virus.

Monoclonal antibodies specific for the gp120 or gp41 portions of the human immunodeficiency virus (HIV) envelope protein gp160 were conjugated to ricin A chain, and their immunotoxic activities against HIV-infected cells were evaluated in the presence or absence of soluble CD4 (sCD4). Immunotoxin activity was measured in vitro as cytotoxicity and inhibition of secretion of infectious HIV. The efficacy of anti-gp41 immunotoxins was enhanced at least 30-fold in the presence of sCD4. This effect was specific for HIV-infected cells, but not for uninfected cells, and was seen at concentrations of sCD4 as low as 0.1 micrograms/ml. Anti-gp120 immunotoxins were marginally inhibited at higher concentrations of sCD4. Flow cytometry analyses showed that sCD4 increased the expression of gp41 on the surface of infected cells and increased internalization of gp120 and gp41. These data suggest that sCD4 alters the cellular trafficking of HIV envelope proteins. These findings also have important implications for the therapeutic use of anti-HIV immunotoxins and may be generalizable to other immunotoxins as well.     

Use of multiple T cell-directed intact ricin immunotoxins for autologous bone marrow transplantation

The monoclonal antibodies (MoAb) T101, G3.7, 35.1, and TA-1 were conjugated to intact ricin using a thioether linkage. These MoAb detect, respectively, the CD5[gp67], CD7[p41], CD2[p50], and [gp95, 170] determinants that are found in the vast majority of cases of T cell acute lymphocytic leukemia (T-ALL). The resulting immunotoxins (ITs) and an equimolar mixture of these ITs were evaluated as potential purgative reagents for autologous transplantation in T-ALL. Leukemic cell lines were used to compare the kinetics of protein synthesis inactivation mediated by each IT. The cells were treated with IT in the presence of lactose in order to block the native binding of ricin. The observed rates of protein synthesis inactivation correlated with target antigen expression detected by fluorescence-activated cell sorter analysis. Of the four ITs, T101-ricin (T101-R) exhibited the fastest rate of inactivation, followed in order by G3.7-ricin, TA-1-ricin, and 35.1-ricin. At concentrations greater than 300 ng/mL, a cocktail containing an equimolar amount of all four ITs (referred to as the four- IT cocktail) exhibited kinetics that were as fast or faster than those of T101-R. The long-term cytotoxic effects of individual ITs and the four-IT cocktail were evaluated using a sensitive clonogenic assay. Each IT was specifically cytotoxic and inhibited 1 to 4 logs of clonogenic leukemic cells at doses (300 to 600 ng/mL) that can be used clinically. The four-IT cocktail was highly cytotoxic; a concentration of 300 ng/mL inhibited greater than 4 logs of leukemic cells while sparing the majority of committed (CFU-GM, CFU-E) and pluripotent (CFU- GEMM) hematopoietic stem cells. The determination of both short-term kinetics of protein synthesis inactivation and longer-term inhibition of clonogenic growth allowed new insight into cell killing by IT. Our results suggest that ITs continue to act on clonogenic target cells for a period of three to five days. Interestingly, the four-IT cocktail was not as potent against clonogenic leukemic cells as T101-R alone, although it exhibited kinetics of protein synthesis inhibition that were as fast as those of T101-R alone. This finding suggests that internalized ITs may differ in the length of time they remain active within the cell. Our results also demonstrate the importance of using several different assays to evaluate IT reagents.     

Adult T cell leukemia: a potential target for ricin A chain immunotoxins

Adult T cell leukemia (ATL) is an almost uniformly fatal malignancy of mature T cells associated with human T cell leukemia/lymphoma virus type 1 (HTLV-1) infection. Cells from this leukemia are characterized by the expression of large numbers of receptors for interleukin 2 (IL- 2). In an attempt to prepare an immunotoxin with selective cytotoxicity for ATL cells, we conjugated anti-Tac, a monoclonal anti-IL-2 receptor antibody, to purified ricin A chains. Although unmodified anti-Tac had no effect on the protein synthesis of these cells, anti-Tac-ricin A chain conjugates produced half-maximal inhibition of protein synthesis in HTLV-1-infected leukemic T cell lines at concentrations of 2 to 6 X 10(-10) mol/L (ID50). An essentially identical ID50 was obtained with leukemic peripheral blood T lymphocytes isolated from two patients with ATL. In contrast, half-maximal inhibition of protein synthesis in HTLV- uninfected, IL-2 receptor-negative T and B cell lines required 200- to 1,000-fold higher concentrations of anti-Tac-ricin A chain conjugates. Both unconjugated anti-Tac and immunoaffinity-purified IL-2 completely inhibited the toxic effects of anti-Tac-ricin A, confirming the specificity of the conjugate-IL-2 receptor interaction. Clonogenic assays demonstrated that anti-Tac-ricin A chain was able to eliminate greater than 99.9% of an HTLV-1-infected T cell population at concentrations only marginally affecting IL-2 receptor-negative cells. The data presented demonstrate that anti-Tac-ricin A is selectively cytotoxic for HTLV-1-infected leukemic T cells in vitro and raises the future possibility of specific therapeutic intervention with immunotoxins in this disease.     

Treatment of murine EL4 leukemia in ascitic form with anti-Thy 1.2 specific immunotoxins

C57BL/6 mice with EL4 leukemia cells in ascitic form were intraperitoneally treated with ricin A chain-multivalent antibody immunotoxins. The immunotoxins containing rabbit IgG anti-Thy 1.2 antibodies complemented by protein A of Staphylococcus aureus were able to interact specifically with the target cells and to induce an antitumor effect as revealed by an increase in survival time of the mice. No apparent secondary effects consecutive to a cytotoxic action on the normal Thy 1.2 antigen bearing cells were observed with the immunotoxin doses used.     

人性化服务在分娩过程中的作用探讨

目的探讨人性化服务方式在分娩过程中的作用。方法将249例产妇随机分为观察组134例和对照组115例。观察组将人性化服务方式贯穿分娩全程,对照组只按常规处理。两组产妇均为无产科合并症,适龄足月初产妇;比较两组的阴道分娩率,剖宫产率,产程、产后出血、新生儿窒息和产妇满意度情况。结果观察组的阴道分娩率明显高于对照组（P〈0．05）,剖宫产率、产后出血率、新生儿窒息率也明显低于对照组（P〈0．05）。总产程及第一、二产程时间均少于对照组（P〈0．05）,观察组的产妇及家属的满意度较对照组有明显提高。结论产时人性化服务可满足产妇的需求,有效的缩短产程,促进阴道分娩,降低剖宫产率,提高产妇的满意度。     

Cell-type-specific cytotoxicity of anti-CD4 and anti-CD8 ricin immunotoxins against human alloreactive T-cell clones

Potent T-cell subset-directed immunotoxins (ITs) were generated by conjugating the anti-CD4 monoclonal antibody (MoAb) G17-2 and the anti- CD8 MoAb G10.1 to the ribosome-inhibitory protein, ricin. The cell-type- specific cytotoxicities of the generated ITs were evaluated at the clonal level using human alloreactive T-cell clones. The kinetics of anti-CD4 ricin-induced inactivation of protein synthesis in target CD4+ cloned T-cells was first order with no detectable lag period and a maximum rate of 0.07 logs per hour (t10 = 13.6 hours; first-order rate constant/K = 0.17 hr-1). The alloantigen specific lytic function of the CD4+ cytolytic T-cell clone JMAC28 was acutely sensitive to anti-CD4 ricin, and no residual lytic activity against allogeneic targets was detectable 24 hours after treatment with as little as 0.5 mmol/L anti- CD4 ricin. Notably, both anti-CD4 ricin and anti-CD8 ricin elicited a selective and dose-dependent inhibition of clonal proliferation of target T-cell clones with a maximum kill of greater than 3 logs at 5 nmol/L. No significant "bystander effects" were observed for non-target cells. Bone marrow progenitor cells CFU-GM, BFU-E, and CFU-GEMM were only minimally affected by either IT. We conclude that these ITs show considerable potential for effective depletion of T-cell subpopulations from allogeneic donor marrow grafts for clinical graft-versus-host disease (GVHD) prophylaxis.     

Induction of caspase-dependent programmed cell death in B-cell chronic lymphocytic leukemia by anti-CD22 immunotoxins

B cells of chronic lymphocytic leukemia (CLL) are long-lived in vivo, possibly because of defects in apoptosis. We investigated BL22, an immunotoxin composed of the Fv portion of an anti-CD22 antibody fused to a 38-kDa Pseudomonas exotoxin-A fragment. B cells from 22 patients with CLL were immunomagnetically enriched (96% purity) and were cultured with BL22 or an immunotoxin that does not recognize hematopoietic cells. The antileukemic activity of BL22 was correlated with CD22 expression, as determined by flow cytometry. BL22 induced caspase-9 and caspase-3 activation, poly(adenosine diphosphate [ADP]-ribose)polymerase (PARP) cleavage, DNA fragmentation, and membrane flipping. Cell death was associated with the loss of mitochondrial membrane potential and the down-regulation of Mcl-1 and X-chromosomal inhibitor of apoptosis protein (XIAP). Furthermore, BL22 induced a proapoptotic 18-kDa Bax protein and conformational changes of Bax. Z-VAD.fmk abrogated apoptosis, confirming that cell death was executed by caspases. Conversely, interleukin-4, a survival factor, inhibited spontaneous death in culture but failed to prevent immunotoxin-induced apoptosis. BL22 cytotoxicity was markedly enhanced when combined with anticancer drugs including vincristine. We also investigated HA22, a newly engineered immunotoxin, in which BL22 residues are mutated to improve target binding. HA22 was more active than BL22. In conclusion, these immunotoxins induce caspase-mediated apoptosis involving mitochondrial damage. Combination with chemotherapy is expected to improve the efficacy of immunotoxin treatment.